init

README.md

@@ -1,3 +1,43 @@
-# Prompt-classification
+# Boosting Meta-Learning for Few-Shot Text Classification via Label-guided Distance Scaling
 
-The code will be released soon.
+## Create python env
+
+```
+conda create -n LDS python=3.7
+source activate LDS
+pip install requirements.txt -r
+```
+
+## Model
+![](./vis/overall.png)
+
+
+## Data
+```
+cd data
+unzip data.zip
+```
+
+## Quick start
+
+```
+cd scripts
+sh ours_all.sh
+```
+The specific parameters in the paper are consistent with ours_all.sh.
+
+**Noting:** before you start, you should download bert-base-uncased from https://huggingface.co/google-bert/bert-base-uncased, and change the path in the ours_all.sh file to your own file path.
+
+## Citation
+
+```
+@article{LIU2025111497,
+title = {Guiding Prototype Networks with label semantics for few-shot text classification},
+journal = {Pattern Recognition},
+volume = {164},
+pages = {111497},
+year = {2025},
+issn = {0031-3203},
+doi = {https://doi.org/10.1016/j.patcog.2025.111497},
+}
+```

prompt_template.csv

@@ -0,0 +1,7 @@
+dataset, template,
+HuffPost,This is a [MASK] news: [sentence],
+Reuters,This is a [MASK] news: [sentence],
+20newsgroup,This is a [MASK] news: [sentence],
+Amazon,This is a [MASK] review: [sentence],
+Clinc150,This is a [MASK] intent: [sentence],
+Banking77,This is a [MASK] intent: [sentence],

requirements.txt

@@ -0,0 +1,8 @@
+matplotlib==3.7.1
+numpy==1.24.3
+pandas==2.0.1
+scikit_learn==1.2.2
+termcolor==2.3.0
+torch==1.11.0+cu113
+tqdm==4.65.0
+transformers==4.28.1